Ascender install on a GCP VM
By popular request, Greg Sowell shows the fastest path to a running Ascender instance on Google Cloud, using a plain Compute Engine VM rather than Google Kubernetes Engine. He creates an e2 instance, raises memory to the recommended baseline of 8 GB with two vCPUs, selects a Rocky Linux 8 boot image optimized for GCP, sets a 30 GB disk, and opens HTTP for the demo, noting that a production deployment should be secured and sized differently.
After the VM boots, he maps its ephemeral public IP to ascender.example.com in his local hosts file, connects over SSH, runs a dnf update, installs git, and clones the Ascender install repository. Instead of editing default.config, he runs the config_vars shell script, which prompts for each option: a new k3s cluster, no offline install, download the kubeconfig, HTTP rather than HTTPS, the local host file, the ascender.example.com hostname, default namespace, admin password and image tag, one replica, and Ledger switched off. The script writes custom.config.yml, and sudo ./setup.sh performs the installation.
A few minutes later the Ascender login page loads at the configured hostname and the admin login succeeds. Viewers evaluating Ascender walk away with the basic sequence for a lab install and a clear list of the settings to change for real use.
Key takeaways
- The recommended baseline for an Ascender VM is 8 GB of RAM and two vCPUs; the default GCP e2 instance needs its memory raised.
- Greg uses a Rocky Linux 8 boot image optimized for GCP with a 30 GB disk, sizing that should grow for production use.
- The demo opens HTTP publicly and maps the ephemeral IP in a hosts file; production needs a persistent IP or DNS name and hardening.
- The config_vars script prompts for every install option, including k3s, kubeconfig download, hostname, namespace and replicas, and writes custom.config.yml.
- Ledger can be skipped during install by choosing option two when the script asks about it.
- Running sudo ./setup.sh from the cloned install directory completes the installation and the Ascender UI loads at the chosen hostname.
Questions this video answers
How do I install Ascender on a Google Cloud VM?
Create a Compute Engine instance with at least 8 GB RAM and two vCPUs on a Rocky Linux 8 image, SSH in, run dnf update, install git and clone the Ascender install repository. Run the config_vars script to answer the setup prompts, which writes custom.config.yml, then run sudo ./setup.sh and log in at your configured hostname.
Do I need Google Kubernetes Engine to run Ascender on GCP?
No. In this demo Greg skips GKE and installs on a regular Compute Engine VM. The config_vars script offers to create a brand new k3s cluster on the VM itself, and the Ascender containers run locally on that single host with one replica.
This video is part of the Ascender Pro playlist. Browse every CIQ video by product and topic.
Transcript
hey everybody I'm Greg Sowell and welcome to another ciq demo today I'm going to be showing you how to get Ascender installed in the Google Cloud as quickly as possible so by popular request on this one um I'm not going to be using G their kubernetes engine I'm just going to be using their regular compute engine I'm going to be installing this on a VM and let me show you all the pieces that will get you there as quickly as possible I am by no means a cloud engineer a cloud expert so your mileage may vary I'm just showing you the most basic
steps required so I'm going to start right in here in Compu engine and VM instances I'm going to click create instance so in here I'm going to come in I'm going to give it a name sender two three 4 why not obviously you want to change your region all that stuff specify it to like meet whatever criteria you want I'm going to leave mine on uh e to for general purpose but I am going to do some customization so by default it's only starting with 4 gigs of RAM I'm going to switch here so that I'll have eight gigs of RAM that's really kind of
our recommended Baseline so you're going to need at least eight with two virtual CPUs we've got that I'm going to move down to uh boot dis I'm going to click change on this and I want to change the OS to some flavor of Rocky Linux whether using the ciq one or just the regular uh rsf one I'm going to go with the rsf I'm going to go for the Linux 8 optimize for gcp and I'm going to change the disc size up to about 30 now obviously if this is your production box right for me this is just demo purposes so I'm just keeping it
really small if it's a production environment you're going to want to crank this up to meet whatever requirements you have so I'm going to say select there for firewalling I'm going to allow HTP through obviously in your real environment you're going to want to secure all this stuff and honestly you're probably not going to want this exposed to the internet for me again I don't care this is only going to live for about 10 minutes so I'm just going to go ahead and expose uh HTTP uh to the internet again you're going to want to customize this to meet your criteria but after that I'm
View full transcriptHide full transcript
going to click create down at the bottom it's going to crank for a few minutes and through the magic of editing you don't have to wait on this we'll come back in just a second when it's up and running so what I'm going to do for this one it starts with an ephemeral Nick and you would probably want to have that as a persistent IP or have a DNS name associated with this box again I'm just doing this quick and dirty for educational purposes so I'm going to copy this I'm going to put this into my host file on my machine here now I'm going to scroll back up to the top on this
box and I'm going to SSH into it right here it's going to pop up this little console allow me to ssh in it'll take just a minute and again through editing you don't see any of this all right now that I have console I'm SSH into this box I'm going to go ahead and do a quick update so I'm going to do sudu dnf update Dy the system will update and when it's done we'll be right back all right now that it's done updating I'm going to install get really quick I'm going to do that sudo dnf install get helps if I spell that correctly
now I'm going to clone the uh install directory that we have created so it's going to be get clone going to paste in the URL it's going to create the new directory so I'm going a CD into Ascender install so in here I have multiple files I can edit the default. config file if I want or I can run through the config vars shell script I haven't actually shown that before in a demo so I'm going to do that really quick do sudu config vars and this is really just kind of a shell script that's going to walk me through and prompt me for
all my options so I'm going to do number one for e or k3s so it's going to ask me if I want a brand new cluster I'm going to say yes so it'll go ahead and install k3s on this box do I want to do offline install nope I'm going to leave that false for now so does the cube config file need to be downloaded I going to say yes for True in My Demo lab I'm just going to be having a HTTP if I was actually doing htps I would need to update the certificate stuff inside of the variables file I'm just going to
leave this default for the routable IP because it's going to be installing the containers on itself the VM on itself so all of it's going to be right there so it can access all that via 1271 I'm just going to tell it true for use the local host file default for where to install the artifacts so this I would actually do some modifications but again for this demo I'm just going to leave all this stuff as is so I'm going to use the Ascender. example.com and in my host file I've actually taken that public IP address that was assigned to it kind of that ephemeral IP and I put that in my host file as asset.
example.com so it's ready to go for me Nam space I'm going to leave that default administrator default password again if this wasn't a demo you would want to change this from default mine's only going to live for about 5 minutes so I'm just going to leave it the default image tag default version default I'm going to go ahead and say true on this one so it'll keep the secrets inside of a CER no external so it's actually doing all this stuff local number of replicat one this is a demo environment I don't need rather I don't need to do anything crazy here my default policy
is if not present and I always put that on all of my execution environments and for this demo I'm not going to install ledger so I'm going to turn it off with number two all right so if I do an LS really quick I can see that it has created a custom config.yml file if I cat that you can see it's filled all this good stuff out and now that I've got all that created I can just execute my shell script for setup so I'm going to do sudu setup.sh and again through the magic of editing you're not have to sit and watch all this we'll come back when it's complete all right so the install is complete despite that little error message that's showing up right there it actually is up and operating I'm going to switch over here to my Ascender.
example.com to do a quick reload and it should pull it up so as you can see it loaded up correctly so I'm going to try and log in admin my admin password and there we have it the system is up and operating so it doesn't really take that much uh you can configure fairly quickly now I went through the uh Editor to uh or rather the shell script to configure that custom uh that Customs configuration file that you could just copy and paste if you've already got it created you can just copy and paste in there again or you can edit the default so it
really comes up and operates pretty rapidly if you have any questions or comments I'd love to hear about them if you can see how you would tweak or tune this please uh as I say give me feedback we always want to hear more from you rather than less so reach out to us again this one was a request from someone in the community so please uh share that stuff happy to build any of the stuff uh if you need some assistance building out your infrastructure we are happy to do that I love uh helping folks out and answering questions we love coming in doing Professional
Services we love uh supporting all this infrastructure give us a shout uh if nothing else thanks and happy automating
Built for scale. Chosen by the world’s best.
2.75M+
Rocky Linux instances
Being used world wide
90%
Of fortune 100 companies
Use CIQ supported technologies
250k
Avg. monthly downloads
Rocky Linux
9
Enterprise products
Spanning the kernel to the orchestrator
Have questions about your infrastructure?
Talk to a CIQ engineer about Rocky Linux, HPC, and AI infrastructure.